JSON vs ParseObject for complex structure

My question is not 100% Parse Server specific, but you have the best overview of the platform, which is why I would like to ask you how you would model the following case:

There is a ParseObject “PdfPage” which contains the binary data of a PDF page (as ParseFile) and some other properties. I would like to save additional metadata for this entity (e.g. list of annotations, dimensions, …) which are currently available to me as JSON.

How would you save this metadata?
a) Save JSON in a string field of “PdfPage” and parse it on the client side.
b) Create a separate ParseObject class (including a mongodb collection) link it using a pointer type.
c) Create a separate ParseObject class (without mogodb collection) and save it in PdfPage.metadata (Type: object).
d) … better suggestion ???

I would be very grateful for your opinion.

Best regards,

I think it depends on the size of this JSON and if you’ll need to execute queries against it. Anyways, you can also consider to create a type Object in your current class.

Many thanks for your response!

The JSON is not large (max. 1000 characters) and it does not have to be searchable.

Type Object is probably the best solution for me. A first test implementation looks very promising. I will use this approach.

Thanks very much!